A Study on the Use of Proxy Reservoir Models in Field Development Optimization and Value of Information Problems

Purwar, Suryansh
Journal Title
Journal ISSN
Volume Title

Mitigating uncertainty in subsurface variables to optimize field development is an intriguing problem still lacking a robust solution. The complexity is further increased by design constraints and quasi-irreversible capital investments. Although versatile optimization techniques and powerful reservoir simulators are available, the primary reason for the lack of a comprehensive solution is the disconnect between optimization and reservoir simulation studies. In this work, we propose an approach that offers a reasonable way to do this. For handling the uncertainties, the proven approach of Experimental Design was used to create cases of possible combination of reservoir variables which were then run in CMG simulator to generate production profiles. Response Surface methodology was employed to generate numerical or ‘proxy’ oil production models from the reservoir simulation results. The numerical models thus generated were plugged into an optimizing model in GAMS to get the desired facility plan. A workflow with these components also allows us to look at the value of buying more information to mitigate uncertainty and the value of incorporating a ‘real’ option of modifying facilities after n years to support initial capital investment decisions